摘要
针对真空感应炉生产过程中温度测量成本较高及精度较差等不足,建立了基于RBF神经网络的真空感应炉终点钢水温度预报模型。对输入参数作了详细的分析、筛选,并运用聚类算法对该模型进行了训练。结合现场数据进行了学习和预报,预报命中率较高,表明采用该方法可很好地预报钢水温度。
A prediction model of molten steel temperature based on RBF neural network was developed to reduce cost and improve temperature control accuracy for vacuum induction melting. The input parameters are described. The network is trained by clustering algorithm. 120 sets of data were used for model building and validation. The experimental results show that the proposed model is effective with high accuracy.
出处
《钢铁研究学报》
CAS
CSCD
北大核心
2005年第4期72-75,共4页
Journal of Iron and Steel Research
基金
国家重大基础研究规划资助项目(2002CB312200)
关键词
真空感应炉
神经网络
温度预报
聚类算法
<Keyword>vacuum induction furnace
neural network
temperature prediction
clustering algorithm